G는 이름이 지정되지 않은 장치에 대한 Android 브라우저 벤치마크를 자랑하며 기자들은 이에 속합니다.

Chrome engineer Eric Seckler, on Google’s Chromium Blog, under the bold headline “Android Sets New Record for Mobile Web Performance”: Today, we are proud to celebrate a major milestone: Android is now the fastest mobile platform for web browsing. Through deep vertical integration across hardware, the Android OS, and the Chrome engine, the latest flagship Android devices are setting new performance records, outperforming all other mobile competitors in the key web performance benchmarks Speedometer and LoadLine and providing a level of responsiveness previously unseen on mobile. Three unnamed Android “flagship phones” scored higher than an unnamed “competing mobile phone platform” (presumably an iPhone 17 Pro) in two benchmarks, Speedometer 3.1 and LoadLine. Speedometer is a longstanding open source benchmark whose development is governed by representatives from WebKit (Apple), Blink (Google), and Gecko (Mozilla). LoadLine is a benchmark from Google and Android OEMs. Speedometer is a benchmark anyone can run just by visiting the benchmark’s website. Running LoadLine, especially on an iOS device, is an enormous hassle that involves two USB-C-to-Ethernet adapters, enabling Remote Automation and the Web Inspector in Safari, installing custom certificates on the iOS device, and installing custom software on an attached Mac. You will be shocked to learn that the three unnamed Android phones outscored the “competing mobile phone” by significantly larger margins on LoadLine than Speedometer. Claiming that these results make Android “the fastest mobile platform for web browsing” is ridiculous. It boggles the mind how many publications parroted Google’s braggadocio — MacRumors, 9to5Google, Android Authority, PhoneArena — without even mentioning that the results can’t possibly be verified because none of the devices (and none of the software versions) are named. This guy at Notebookcheck even had the audacity to put in his headline that Google “shows the receipts” for its claims. What kind of receipt doesn’t say what you bought? I would love to wager real money with the authors of any of those stories on what the Speedometer 3.1 results show for 100 random real-world Android users vs. 100 random real-world iPhone users. Or how about the average scores from the three best-selling models on each platform from the last year. Name the devices or shut up. ★ Thursday, 26 March 2026